The explosive growth in mobile services imposes much pressure on operators in network expansion. In order to reduce the bandwidth consumption of transmission networks of the core network and the radio access network (RAN, Radio Access Network), reduce operating costs of services, and implement differentiated quality of service (QoS, Quality of Service), network are evolving toward flat structures.
In a conventional radio communication system, all service data is cached in a base station controller, and a specific process of scheduling the service data is as follows:
A base station controller computes a user priority for each user according to information such as service types and service priorities of all services of each user; the base station controller sends the computed user priority to a base station; the base station authorizes each user (that is, grants an air interface bandwidth to each user) according to the user priority and the air interface channel environment, and returns an authorization result to the base station controller; the base station controller authorizes the services of each user (that is, grants an air interface bandwidth to each service) based on an authorization state of the user, and sends service data of the user to the base station according to an authorization result; and the base station sends the service data to a user equipment (UE, User Equipment) over the air interface.
However, in order to implement network expansion, a local breakout (LBO, Local Breakout) function is implemented on a base station at present, that is, most low-priority services are directly distributed from the base station to an Ethernet. In such a scenario, service data is not completely cached in a base station controller, and service data of one user may be partially cached in a base station controller, and partially cached in a base station. It can be seen that the existing service data scheduling method is not suitable for the scenario, and cannot solve the service data scheduling problem in such a scenario.